Screening for Highly Transduced Genes in Staphylococcus Aureus Reveals Both Lateral 2 and Specialized Transduction

Home , Prophage

bioRxiv preprint doi: https://doi.org/10.1101/2020.10.28.360172; this version posted October 29, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC 4.0 International license.

1 Screening for highly transduced genes in Staphylococcus aureus reveals both lateral 2 and specialized transduction

4 Janine Zara Bowring1, Yue Su1, Ahlam Alsaadi1, Sine L. Svenningsen 2, Julian Parkhill3 & Hanne 5 Ingmer1*

7 1Department of Veterinary and Animal Sciences, University of Copenhagen

8 2Department of Biology, University of Copenhagen

9 3Department of Veterinary Medicine, University of Cambridge

11 * Corresponding author: Hanne Ingmer, [email protected]

1 bioRxiv preprint doi: https://doi.org/10.1101/2020.10.28.360172; this version posted October 29, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC 4.0 International license.

15 Abstract

16 Bacteriophage-mediated transduction of bacterial DNA is a major route of horizontal gene transfer in the 17 human pathogen, Staphylococcus aureus. Transduction involves packaging of bacterial DNA by viruses and 18 enables transmission of virulence and resistance genes between cells. To learn more about transduction in S. 19 aureus, we searched a transposon mutant library for genes and mutations that enhanced transfer mediated by 20 the temperate phage, ϕ11. Using a novel screening strategy, we performed multiple rounds of transduction of 21 transposon mutant pools selecting for an antibiotic resistance marker within the transposon element. When 22 determining the locations of transferred mutations, we found that, within each pool of 96 mutants the screen 23 had selected for just 1 or 2 transposon mutant(s). Subsequent analysis showed that the position of the 24 transposon, rather than inactivation of bacterial genes, was responsible for the phenotype. Interestingly, from 25 multiple rounds we identified a pattern of transduction that encompassed mobile genetic elements, as well as 26 chromosomal regions both upstream and downstream of the phage integration site. The latter was confirmed 27 by DNA sequencing of purified phage lysates. Importantly, transduction frequencies were lower for phage 28 lysates obtained by phage infection rather than induction. Our results confirm previous reports of lateral 29 transduction of bacterial DNA downstream of the integrated phage, but also indicate specialized transduction 30 of DNA upstream of the phage, likely involving imprecise excision of the phage from the bacterial genome. 31 These findings illustrate the complexity of transduction processes and increase our understanding of the 32 mechanisms by which phages transfer bacterial DNA.

2 bioRxiv preprint doi: https://doi.org/10.1101/2020.10.28.360172; this version posted October 29, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC 4.0 International license.

42 Importance

43 Horizontal transfer of DNA between bacterial cells contributes to the spread of virulence and antibiotic 44 resistance genes in human pathogens. For Staphylococcus aureus, bacterial viruses are particularly important. 45 These viruses, termed bacteriophages, can transfer bacterial DNA between cells by a process known as 46 transduction, which despite of its importance is only poorly characterized. Here, we employed a transposon 47 mutant library to investigate transduction in S. aureus. We show that the location of bacterial DNA in relation 48 to bacteriophages integrated in the bacterial genome is a key decider of how frequently that DNA is 49 transduced. Based on serial transduction of transposon mutant pools and direct sequencing of bacterial DNA in 50 bacteriophage particles, we demonstrate both lateral and specialized transduction. The use of mutant libraries 51 to investigate the patterns of bacterial DNA transfer between cells could help understand how bacteria evolve 52 virulence and resistance and may ultimately lead to new intervention strategies.

3 bioRxiv preprint doi: https://doi.org/10.1101/2020.10.28.360172; this version posted October 29, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC 4.0 International license.

67 Introduction

68 S. aureus is a gram-positive opportunistic pathogen that causes a wide-range of disease in humans. It can 69 become resistant to a variety of antibiotics like methicillin (MRSA), tetracycline and vancomycin, and resistance 70 is limiting treatment options (1, 2) . The staphylococcal genome is highly plastic, largely due to horizontal gene 71 transfer mediated by mobile genetic elements (MGEs) such as staphylococcal pathogenicity islands (SaPIs), 72 plasmids and phages (3, 4). Particularly, phage mediated transduction appears to be an important route of 73 gene transfer. Transduction is a process whereby some phages are able to package bacterial DNA at low 74 frequency and transfer this DNA between cells (5, 6). Transducing phages are often temperate as they can both 75 be integrated in the bacterial genome as prophages (lysogeny) or be lytic, where they replicate before lysing 76 the cell and releasing the progeny phages. Transduction can occur through a number of different mechanisms; 77 generalised transduction, specialised transduction, and lateral transduction. In S. aureus, generalized 78 transduction has been used for many years as a genetic tool and recently, the mechanism for lateral 79 transduction was established (7, 8). In contrast we know very little about specialized transduction in this 80 organism.

81 The often-used S. aureus laboratory strain 8325-4 is the phage-cured equivalent of strain 8325, which contains 82 three prophages (ϕ11, ϕ12, and ϕ13) ((9). Phage ϕ11 is one of the best characterised of the staphylococcal 83 phages, and regularly used as a laboratory tool for transferring genes by transduction (10, 11). As a prophage, it

84 sits in an intergenic region of the 8325 chromosome with a specific directionality, where the attL is situated at

85 ~1.967 Mb and the attR at ~1.923 Mb (12, 13). Phage ϕ11 is a pac type phage, meaning that it recognises a 86 phage ‘pac’ site and packages DNA into the capsid in a ‘headful’ manner, terminating when the capsid is filled 87 with slightly more than the ~45 kb phage genome (14). As such, ϕ11 can transfer bacterial DNA by generalised 88 transduction, which occurs when bacterial DNA contains sequences that have homology to the phage ‘pac’ 89 sites and so are recognised by the phage packaging machinery (14). Generalised transduction can occur either 90 when a phage infects a bacterium and enters the lytic cycle, or when a resident prophage is induced.

91 Lateral transduction is the most recently identified form of phage-mediated transduction in S. aureus (14). 92 Previously, it was thought that prophage induction initiated the excision, replication, and packaging cycle in 93 that order, with replication of the phage only occurring post-excision. However, it has since been determined 94 that in S. aureus some phages begin replication before excising from the host chromosome (14). This leads to 95 the replication of the integrated prophage together with the flanking regions of bacterial DNA. Because the 96 flanking bacterial DNA is replicated with the integrated prophage, the bacterial DNA downstream of the pac

4 bioRxiv preprint doi: https://doi.org/10.1101/2020.10.28.360172; this version posted October 29, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC 4.0 International license.

97 site will be packaged by the headful mechanism. In practice, this has led to transduction frequencies of 1000- 98 fold higher rates when compared to generalised transduction rates and to higher rates of transduction up to 99 100 kb downstream of the prophage. The study identified that this effect was unidirectional from the pac site 100 and so the region upstream of the phage was not transferred (15). Lateral transduction has only been shown to 101 occur with pac prophages that have been induced and does not occur when a phage infects and undergoes the 102 lytic cycle (16, 17).

103 Specialised transduction occurs when a prophage excises incorrectly from the bacterial chromosome (16–24). 104 Instead of a precise excision of the full phage genome between the attL and attR sites formed at integration, an 105 incomplete portion of the phage excises alongside some of the flanking bacterial DNA both upstream and 106 downstream of the phage. As long as the phage pac site is present, this hybrid DNA will be replicated, packaged 107 into phage capsids, and transduced (15, 17, 20). Specialised transduction only occurs when a prophage is 108 induced, rather than when a phage infects and undergoes the lytic cycle. It is limited to the immediate regions 109 of bacterial DNA flanking a prophage. Specialised transduction mechanisms have been reported for phages 110 infecting a number of bacterial genera, including Bacillus, Salmonella, Pseudomonas and Vibrio (14, 25). 111 However, there is an absence of literature on specialised transduction by staphylococcal phages, with most 112 ϕ11 papers referring to the work on Salmonella phage P22 when describing the process of specialised 113 transduction (25).

114 Different bacteria and the prophages display different patterns of transduction (26). This means that different 115 strains within the same species, but that contain differing prophages, can have varying patterns of 116 transduction. A recent bioRxiv preprint described the use of DNA short- and long-read sequencing to 117 investigate the patterns of bacterial DNA in transductant particles between different genera and species of 118 bacteria (27). This pattern could be explained by different phages and prophages being more or less prone to 119 lateral, specialised or generalised transduction and by the number of pseudo-pac sites in the strain’s 120 chromosome. Furthermore, the same temperate phage may transduce bacterial DNA differently depending on 121 whether that phage is undergoing lytic infection or lysogenic induction. Both lateral and specialized 122 transduction require the induction of an integrated prophage, as both are initially reliant on the phage being 123 integrated into the bacterial chromosome. Generalised transduction, on the other hand, can occur during both 124 life cycles.

125 To further understand transduction in S. aureus, we explored a screening methodology that allowed us to 126 identify S. aureus transposon mutants that were preferentially transduced from a pool of mutants. We utilised

5 bioRxiv preprint doi: https://doi.org/10.1101/2020.10.28.360172; this version posted October 29, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC 4.0 International license.

127 the Nebraska transposon mutant library (NTML) in S. aureus strain JE2, infecting pools of 96 mutants with the 128 transducing, lysogenic phage ϕ11 (27). The resultant lysates were transduced to recipient strain 8325-4 ϕ11, 129 from which pooled colonies could be induced by mitomycin C and transduced in cycles, selecting for 130 transductants with the transposon-encoded resistance marker. From this screen we anticipated identifying 131 genes that either, when inactivated, promoted transduction, or were located within the chromosome in 132 regions transduced at higher rates than elsewhere. Our results revealed location-specific effects of the highly 133 transduced regions flanking the phage ϕ11 integration site, by comparing the transduction frequency of other 134 genes located in the same region. Indeed, the regions flanking the ϕ11 integration site were further confirmed 135 as preferentially transduced regions of bacterial DNA by sequencing purified phage lysates.

136 Here, we show that the transposon mutants in the chromosomal regions were transduced by both lateral and 137 specialised transduction. This study establishes a simple screening method that reveals patterns of 138 transduction at the genome-level, which can be readily applied to transposon mutant libraries of other 139 bacteria. Remarkably, this study is the first to demonstrate specialised transduction in S. aureus.

140 Materials and Methods

141 Bacterial strains and growth conditions

142 Bacterial strains used in this study are listed in supplementary table 1. S. aureus strains were grown in TSB and 143 on TSA plates (Oxoid), with or without antibiotics as appropriate.

144 Bacteriophage titre

145 Phage titre quantifications were performed essentially as previously described (27). Briefly, recipient strains

146 were grown to 0.35 OD600 and 100 μL aliquots of recipient mixed with 100 μL phage lysate, at different

0 -8 147 dilutions in phage buffer (10 – 10 ; MgSO4 1 mM, CaCl2 4 mM, Tris-HCl pH 8 50 mM, NaCl2 0.1 M). After 10 148 minutes incubation at room temperature, 3 mL of liquid PTA (phage top agar; Oxoid nutrient broth n°2, agar 149 3.5% w/v) was added and the mixture poured out on PB plates (phage base; nutrient broth n°2, agar 7% w/v). 150 Plates were incubated at 37°C overnight and plaques counted.

151 Bacteriophage transduction

152 Phage transductions were performed as previously described (27). Briefly, recipient strains were grown to 1.4

153 OD600, 4.4 mM CaCl2 added and 1 mL aliquots of recipient mixed with 100 μL phage lysate, at different dilutions 154 in phage buffer (100 – 10-4). After 20 minutes incubation at 37°C, 3 mL of liquid TTA (transduction top agar; TSA,

6 bioRxiv preprint doi: https://doi.org/10.1101/2020.10.28.360172; this version posted October 29, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC 4.0 International license.

155 50% agar) was added and the mix poured out on selective TSA plates. Plates were incubated at 37°C for 24 156 hours and colonies counted.

157 Bacteriophage infection and induction

158 Bacteriophage infection was performed as previously described (26, 28). Briefly, recipient bacteria were grown

159 to 0.15 OD600 and centrifuged, before re-suspending the pellet in 1:1 TSB and phage buffer. The culture was 160 then infected with phage at an MOI of 1 and incubated at 30°C, 80 rpm until complete visual lysis, normally 161 after 3 – 4 hours. Lysates were then filtered using 0.22 μM filters and stored at 4°C.

162 Bacteriophage induction was performed as previously described (29). Briefly, lysogens were grown to 0.1 – 0.2

-1 163 OD600 and 2 μg ml mitomycin C (Sigma, from Streptomyces caespitosus) added to induce resident prophages. 164 Cultures were then incubated at 30°C, 80 rpm until complete visual lysis, normally after 3 – 4 hours. Lysates 165 were filtered using 0.22 μM filters and stored at 4°C.

166 NTML transposon transduction screening

167 For screening plates from the NTML, one 96-well plate was replicated and grown overnight in selective TSB 168 (erythromycin, 10 μg ml-1) before pooling 10 μL of each mutant in 100 mL selective TSB, in a 500 mL flask. The 169 culture was infected with ϕ11 (MOI 1) as described. The resulting lysate was then used to transduce the NTML 170 transposons to 8325-4 ϕ11 recipient, as described. The plate resulting from the transduction of non-diluted 171 lysate was harvested, pooled and grown in selective TSB and the culture was diluted then induced. The 172 resulting lysate was used to transduce into the 8325-4 ϕ11 recipient a second time, and the transductants 173 were pooled and induced. From the resulting lysate, a third transduction was performed, 10 transductant 174 colonies selected and the location of the transposon mutant identified as previously described (30).

175 DNA sequencing

176 Preparation of the phage lysates for DNA sequencing was done by precipitating the phage capsids before DNA 177 extraction. For the phage precipitation, DNase (2.5 U ml-1, Thermo) and RNase (1 μg ml-1, Sigma) were added to

178 the filtered lysate before incubation at 37°C for 1 hour. NaCl2 (1M) was added and mixed on ice with an orbital 179 shaker at 4°C for 1 hour. The lysate was then centrifuged at 11,000 x g for 10 minutes at 4°C and supernatant 180 collected and mixed well with PEG 8000 (10% w/v) before incubation on ice overnight at 4°C. The lysate was 181 then centrifuged at 11,000 x g for 10 minutes at 4°C and the supernatant discarded, with falcon tubes inverted 182 and dried. The pellet appears as two vertical white lines and was resuspended in 8 mL phage buffer per 500 mL

7 bioRxiv preprint doi: https://doi.org/10.1101/2020.10.28.360172; this version posted October 29, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC 4.0 International license.

183 lysate, with resuspension for 1 hour at 4°C. DNA extraction was then performed using the MagAttract HMW 184 DNA Kit (Qiagen) as per manufacturer’s instructions. DNA concentration and absorbance peaks were checked 185 using a Thermo NanoDrop 2000 and a Qubit Flex Fluorometer. A draft genome for 8325-4 was generated by 186 sequencing chromosomal DNA on an Illumina miSeq following the manufacturer’s instructions. The sequence 187 was assembled, and the contigs ordered using the Sanger Institute pipelines (31). Phage lysate DNA was 188 sequenced in the same way, and the resulting reads were mapped to this genome using SMALT (26, 28) and 189 visualised using Artemis (26, 28).

190 DNA methods

191 General DNA manipulations were performed using standard procedures. Primers used in this study can be 192 found in supplementary table 2. PCRs were performed using DreamTaq Green master mix (2x; Thermo) as per 193 manufacturer’s instructions. PCRs for identification of transposon locations were performed as previously 194 described (7, 9). PCR products were sequenced at Eurofins MWG Operon.

195 For identification of the presence of prophages in strains, a multiplex PCR for the main integrases of the 196 staphylococcal phages was performed, as previously described (32). Initial screening of colonies was performed

197 using 1 μL of template from a single colony resuspended in 50 μL ddH2O, boiled for 10 mins and on ice for 10 198 mins. Results were confirmed using DNA template (10 ng), extracted using GenElute™ Bacterial Genomic DNA 199 Kit as per manufacturer’s instructions (Sigma-Aldrich).

200 For the long amplifications necessary for amplifying regions of specialised transduction, the Takara LA 201 polymerase PCR kit was used as per manufacturer’s instructions with the following alteration, a 30 s annealing 202 step was added at 55°C. Templates used were either bacterial DNA extracted with the GenElute™ Bacterial 203 Genomic DNA Kit or phage DNA extracted from phage lysates that had been precipitated as previously 204 described and then extracted using the GenElute™ Bacterial Genomic DNA Kit with an extended proteinase K 205 incubation of 1 hour (Sigma-Aldrich). 50 ng of template was used for each PCR. The resulting products were run 206 out on a 0.7% agarose gel at 30 V overnight.

207 Statistical analyses

208 Data analyses were performed in GraphPad Prism 8 software. Specific statistical tests are indicated in the figure 209 legends, where appropriate.

210

8 bioRxiv preprint doi: https://doi.org/10.1101/2020.10.28.360172; this version posted October 29, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC 4.0 International license.

211 Results

212 Transduction selection preferentially identifies transposon mutants in MGEs and regions flanking prophage 213 ϕ11

214 With the aim of identifying genes or mutations that increase transduction frequency, we screened the S. 215 aureus JE2 NTML for mutations that were preferentially transduced. To achieve this, we carried out 216 consecutive rounds of transduction on pools of library mutants by combining 96-well plates of transposon 217 mutants and infecting these with the transducing temperate phage ϕ11, to produce an ‘infection’ phage lysate 218 containing transducing particles. We then performed transduction using these lysates, with the laboratory 219 strain 8325-4 ϕ11 as recipient. Having the ϕ11 prophage in the recipient prevents phage superinfection and 220 thus increases survival of transductants. Furthermore, previous studies have shown that lysogenic strains can 221 undergo autotransduction, where a lysogen spontaneously releases phages that infect nearby susceptible 222 strains, resulting in transductants that can return useful bacterial DNA to the original lysogenic population (27, 223 33). Transductants carrying the transposon element were selected for by their resistance to erythromycin, 224 which is encoded within the transposon. These transductants were then pooled and the ϕ11 prophage was 225 induced to produce a ‘round 1 induction’ lysate. The transduction and induction steps were repeated to 226 produce ‘round 2’ and ‘round 3’ transduction lysates. At each stage the CFU ml-1 and PFU ml-1 were recorded 227 and the transduction frequency was calculated, controlling for phage variation (CFU ml-1/ PFU ml-1; Fig 1). 228 Importantly, the transduction frequencies showed a general increase over the three rounds of transduction 229 screening for the eight 96-well plates examined, demonstrating that our selection procedure had succeeded 230 (Fig 1).

9 bioRxiv preprint doi: https://doi.org/10.1101/2020.10.28.360172; this version posted October 29, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC 4.0 International license.

231

232 Figure 1. The transduction frequencies of pooled transposon mutants. Transduction frequencies (log 233 transformed CFU ml-1/PFU ml-1) for the initial infection transduction (white), the rounds 1 (light grey), 2 (dark 234 grey), and 3 (black) induction-transduction. CFU ml-1 was calculated from the number of transductant colonies 235 on erythromycin selective plates and PFU ml-1 from the number of plaques formed on a susceptible recipient 236 (8325-4). Each bar represents one replicate of a pool of 96 transposon mutants labelled by library plate number 237 (3.1 – 4.4).

238

239 At the round 3 transduction, we collected 10 different colonies from each plate pool and identified the 240 locations of the transposon mutations by digestion, ligation, and sequencing of the transposon-chromosomal 241 flanks, using an established protocol (14). Examination of the 10 colonies from each of the different pools 242 revealed that in the third round transductants, most pools contained just one predominant transposon mutant 243 (Fig 2). For 7 of the 8 pools, only one transposon mutant was identified in the final round of colony check, 244 whilst for plate 3-1, 2 transposon mutants were identified. The first striking observation was that the majority 245 (5 out of 9) of the transposon mutants identified in our screen were situated on MGEs (Fig 2). Of these, one 246 transposon was located in a SaPI5 gene, one in the USA300ϕSa2 prophage of JE2, while the other 3 were 247 located in the USA300ϕSa3 JE2 prophage. These MGE-located transposon mutants were distributed

248 throughout the chromosome (Fig 2). When we identified the locations of the other 4 chromosomally situated 249 non-MGE related transposon mutants, we found that these clustered around the integration site of the ϕ11

250 (attB; Fig 2). Relative to the ϕ11 prophage directionality, 2 of the transposon mutants were situated upstream,

10 bioRxiv preprint doi: https://doi.org/10.1101/2020.10.28.360172; this version posted October 29, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC 4.0 International license.

251 but close to, the ϕ11 attB, whilst the other 2 were located downstream of this site (9, 14). We hypothesised 252 that we were identifying transposon mutants that were preferentially transduced due to either induced MGE

253 mobility or lateral and specialised transduction.

254

11 bioRxiv preprint doi: https://doi.org/10.1101/2020.10.28.360172; this version posted October 29, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC 4.0 International license.

255 Figure 2. Locations of the identified transposon mutants in the JE2 genome. Depiction of NTML JE2 genome 256 (NCBI accession number 007793.1), with MGEs USA300ϕSa2 (pink arrow), USA300ϕSa3 (blue arrow), and JE2

257 SaPI5 (purple arrow). The ϕ11 insertion site (attB) is indicated. The genes identified with a transposon mutant 258 insertion are marked with the NTML gene identification number, as is the position of the Tn_rotB control. The 259 table identifies the gene in which the transposon was inserted for each identified mutant. For all plates this 260 was either 1 or 2 mutants. ‘Plate’ refers to the NTML 96-plate numbering of the pooled 96 mutants, ‘location’ 261 identifies whether the transposon insertion was in a MGE-situated gene or in the chromosome (Chr). Upstream 262 and downstream refer to the position of the gene relative to the ϕ11 prophage insertion, using the prophage

263 directionality (eg. downstream = downstream of the attR, upstream = upstream of the attL).

264

265 Transposon mutants up- and down-stream of ϕ11 have an increased transduction frequency

266 The chromosomal transposon mutants clustering around the ϕ11 attB interested us as potential incidents of 267 specialised or lateral transduction (SAUSA300_1866, SAUSA300_1868, SAUSA300_1849, and SAUSA300_1797). 268 Therefore, we transduced these individual transposon mutants from JE2 into the clean laboratory strain 8325-4 269 ϕ11, which has only the ϕ11 prophage, and confirmed by "integrase PCR" that none of the JE2 endogenous 270 phages had been transferred. Using these ‘clean background’ strains, we confirmed that the transduction 271 frequencies in the 4 chromosomal transposon mutants were higher than our 8325-4 ϕ11 rotB::erm transposon 272 control. The rotB transposon mutant has a wild-type phenotype in respect to bacterial growth and 273 transduction, and ϕ11 should only transfer this transposon mutant by generalised transduction, based on its 274 position in the 8325 genome (1.794 Mb, figure 2). All 4 transposon mutants had a higher transduction 275 frequency than the rotB control, with mutants SAUSA300_1868, SAUSA300_1866, SAUSA300_1849, and 276 SAUSA300_1797 having significantly higher frequencies (Fig 3).

12 bioRxiv preprint doi: https://doi.org/10.1101/2020.10.28.360172; this version posted October 29, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC 4.0 International license.

277

278 Figure 3. The CFU ml-1, PFU ml-1, and CFU/PFU transduction frequency of the transposon mutants located up- 279 and downstream of the ϕ11 integration site. A shows the CFU ml-1 (black) and PFU ml-1 (grey) for control 280 strains 8325-4 ϕ11 and 8325-4 ϕ11 Tn_rotB with the 4 different Tn mutants up- and downstream of the ϕ11 281 integration site (Tn inserted into SAUSA300_1868, SAUSA300_1866, SAUSA300_1849, and SAUSA300_1797 282 respectively). The values shown have been log transformed and show the mean and standard deviation (SD) for 283 4 biological replicates. B shows the CFU/PFU transduction frequency for each mutant compared with Tn_rotB 284 (mean and SD, 4 biological replicates). Unpaired t tests were performed and p-values from left to right were 285 0.0312*, 0.0346*, 0.0008***, <0.0001****.

286 Transposon mutants downstream of the ϕ11 integration site are transferred by lateral transduction

287 Next was to further define the type of transduction occurring with these 4 chromosomal transposon mutants. 288 Specialized transduction occurs in the immediate flanks neighbouring the prophage, so it was not a likely 289 explanation for Tn_1797. Lateral transduction occurs in bacterial DNA up to 100 kb downstream of a laterally 290 transducing prophage, based on the pac site encoded by that phage. In the case of ϕ11, the pac site is located

291 in the small terminase gene and this means the region downstream of the attR (<1.923 Mb) is transferred by

13 bioRxiv preprint doi: https://doi.org/10.1101/2020.10.28.360172; this version posted October 29, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC 4.0 International license.

292 lateral transfer, whilst the region upstream of the attL (>1.967 Mb) is not (14). To establish whether the higher 293 transduction frequencies seen in the mutants with transposons located downstream of ϕ11 were due to the 294 effects of the specific transposon insertions or the more general effect of lateral transduction, we tested the 295 transduction frequency of other transposon mutants from the same region of DNA. We used three different 296 mutants where the transposon was situated close to the downstream transposon insertion sites identified from 297 our pooled screens, but had not been identified by the screen (Tn_1799, Tn_1843, and Tn_1852). We 298 transduced these transposon mutants into the 8325-4 ϕ11 background and measured the transduction 299 frequency following induction with mitomycin C. All three of the transposon mutants had a significantly higher 300 transduction frequency than the Tn_rotB control (Fig 4), and were comparable to the Tn_1849 and Tn_1797 301 mutants. That these mutants also had a higher transduction frequency similar to the mutants from the screen, 302 suggested that the cause for these higher rates was lateral transduction. Since lateral transduction only occurs 303 when a prophage is induced, we tested the transduction frequency of both the original mutants and these 304 ‘neighbouring’ mutants in an infection set up, where lateral transduction could not occur.

305

306 Figure 4. The CFU ml-1, PFU ml-1, and CFU/PFU transduction frequency of transposon mutants from the 307 chromosomal region downstream of the ϕ11 integration site that were not selected in the original 308 transduction screen. A shows the CFU ml-1 (black) and PFU ml-1 (grey) for the controls 8325-4 ϕ11 and 8325-4 309 ϕ11 Tn_rotB with 3 different Tn mutants inserted separately in genes downstream of the ϕ11 integration site 310 (Tn inserted into SAUSA300_1852, SAUSA300_1843, and SAUSA300_1799 respectively). The values shown have

14 bioRxiv preprint doi: https://doi.org/10.1101/2020.10.28.360172; this version posted October 29, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC 4.0 International license.

311 been log transformed and show the means and standard deviation for 4 biological replicates. B shows the 312 CFU/PFU transduction frequency (mean and SD, 4 replicates) for each mutant compared with Tn_rotB. A one- 313 way ANOVA was performed with Dunnetts multiple comparison test comparing the mean of the Tn_rotB 314 control with each mutants mean, and p-values from left to right were 0.0006***, 0.0203*, and 0.0010**.

315

316 To measure the transduction frequency following infection rather than induction, the transposon mutants in 317 the original JE2 library background were infected with ϕ11 and the transduction frequency of the resulting 318 lysate was established. The transduction frequencies of all the transposon elements located downstream of 319 ϕ11were higher than the Tn_rotB control (Fig 5A-D); however, these transduction frequencies were reduced 320 when compared to the previous induction-linked frequencies (Fig 5E-G). This supported the idea that lateral 321 transduction was the main cause of the increased transduction frequencies downstream of the ϕ11 integration 322 site.

15 bioRxiv preprint doi: https://doi.org/10.1101/2020.10.28.360172; this version posted October 29, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC 4.0 International license.

323

324 Figure 5. The CFU ml-1, PFU ml-1, and CFU/PFU transduction frequency of downstream JE2 transposon 325 mutants following infection with ϕ11. A shows the CFU ml-1 and PFU ml-1 for the control JE2 Tn_rotB and the 326 different screen identified Tn mutants downstream of the ϕ11 integration site following infection with ϕ11 (Tn 327 inserted into SAUSA300_1797 and SAUSA300_1849 respectively). B shows the CFU/PFU transduction 328 frequency for each mutant compared with Tn_rotB. P-values for a one-way ANOVA with Dunnetts multiple

16 bioRxiv preprint doi: https://doi.org/10.1101/2020.10.28.360172; this version posted October 29, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC 4.0 International license.

329 comparison test were 0.0003*** and 0.0027** (left to right). C shows the CFU ml-1 and PFU ml-1 for the control 330 JE2 Tn_rotB and the other Tn mutants downstream of the ϕ11 integration site following infection with ϕ11 (Tn 331 inserted into SAUSA300_1799, SAUSA300_1843 and SAUSA300_1852 respectively. D shows the CFU/PFU 332 transduction frequency for each mutant compared with Tn_rotB. P-values for a one-way ANOVA with Dunnetts 333 multiple comparison test were 0.0233*, 0.0476*, and 0.0233* (left to right). All values shown have been log 334 transformed and show the means and SD from the 3 replicate values. E-G, the CFU/PFU transduction 335 frequency following either infection or induction with ϕ11. E compares the induction (square) and infection 336 (circle) transduction frequencies of the neighbouring transposon mutants and the Tn_rotB control, showing the 337 means and SD. P-values for a two-way ANOVA with Sidak multiple comparison test were 0.9944ns, 0.8400ns, 338 0.0212* and 0.0190* left to right. F and G compare the induction (square) and infection (circle) transduction 339 frequencies of the Tn_rotB control together with Tn_1797 and Tn_1849, respectively. P-values for a two-way 340 ANOVA with Sidak multiple comparison test were (F) 0.0830ns and <0.0001**** (G) 0.0874ns and 0.0747ns.

341

342 Transposon mutants upstream of the ϕ11 integration site are transferred by specialised transduction

343 To establish whether the genes located upstream and close to the ϕ11 integration site were transduced at 344 higher frequencies due to gene specific effects or because of specialised transduction, additional transposon 345 mutants from this region were identified and tested. The mutants in the same region as Tn_1866 and Tn_1868 346 showed comparable transduction frequencies, whilst the mutant furthest away from the phage integration site 347 showed a gradual reduction in transduction frequency, albeit higher than the Tn_rotB control (Fig 6. Tn_1865, 348 Tn_1883 and Tn_1898 vs Tn_1916, respectively).

17 bioRxiv preprint doi: https://doi.org/10.1101/2020.10.28.360172; this version posted October 29, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC 4.0 International license.

349

350 Figure 6. The CFU ml-1, PFU ml-1, and CFU/PFU transduction frequency of transposon mutants from the 351 chromosomal region upstream of the ϕ11 integration site that were not selected in the original transduction 352 screen. A, shows the CFU ml-1 and PFU ml-1 for the controls 8325-4 ϕ11 and 8325-4 ϕ11 Tn_rotB with 4 353 different Tn mutants inserted separately in genes upstream of the ϕ11 integration site (Tn inserted into 354 SAUSA300_1865, SAUSA300_1883, SAUSA300_1898 and SAUSA300_1916 respectively, progressively getting 355 further away from the ϕ11 insertion site from left to right). B, shows the CFU/PFU transduction frequency for 356 each mutant compared with Tn_rotB. A one-way ANOVA was performed with Tukey multiple comparison test 357 comparing the means. Comparison of the Tn_rotB control with each mutants mean gave p-values from left to 358 right of <0.0001 ***, 0.0005 ***, 0.0006***, and 0.0022*. An additional comparison of the Tn mutant closest to 359 the ϕ11 insertion site (Tn_1865) to the one furthest away (Tn_1916) is indicated by the black line, with a p- 360 value of 0.0188*. All values shown have been log transformed and show the means and SD for 3 biological 361 replicates.

362

363 These results suggested that specialised transduction was occurring in the region immediately upstream of the 364 ϕ11 and causing the higher levels of transduction seen. Furthermore, when the equivalent JE2 transposon 365 mutants were tested following ϕ11 infection, the transduction frequencies were lower and not significantly 366 different to the Tn_rotB control (Fig 7). Given that specialised transduction requires integration of the phage

18 bioRxiv preprint doi: https://doi.org/10.1101/2020.10.28.360172; this version posted October 29, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC 4.0 International license.

367 genome into the host chromosome, which isn’t thought to occur during infection, this further supports that the 368 upstream transposon mutants were being identified due to the occurrence of specialised transduction. Indeed, 369 by using primers annealing within the NTML transposon (Martn-ermR) and the ϕ11 phage (phi11-direction-3f) 370 it was possible to amplify DNA products from induced and precipitated lysates of 8325-4 ϕ11 Tn_1866 and 371 8325-4 ϕ11 Tn_1868 (Fig. 8). This indicated that some particles contained the upstream regions of bacterial 372 DNA alongside the N-terminal of the phage genome, as would be expected in cases of specialised transduction.

373

374 Figure 7. The CFU ml-1, PFU ml-1, and CFU/PFU transduction frequency of upstream JE2 transposon mutants 375 following infection with ϕ11. A, shows the CFU ml-1 and PFU ml-1 for the control JE2 Tn_rotB and the different 376 screen-identified Tn mutants upstream of the ϕ11 integration site following infection with ϕ11 (Tn inserted

19 bioRxiv preprint doi: https://doi.org/10.1101/2020.10.28.360172; this version posted October 29, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC 4.0 International license.

377 into SAUSA300_1866 and SAUSA300_1868 respectively). B, shows the CFU/PFU transduction frequency for 378 these Tn mutants compared with Tn_rotB. P-values for a one-way ANOVA with Dunnetts multiple comparison

ns ns -1 -1 379 test were 0.5807 and 0.2225 . C, shows the CFU ml and PFU ml for the control JE2 Tn_rotB and the 380 different neighbouring Tn mutants upstream of the ϕ11 integration site following infection with ϕ11 (Tn 381 inserted into SAUSA300_1865, SAUSA300_1883, SAUSA300_1898 and SAUSA300_1916 respectively). D, shows 382 the CFU/PFU transduction frequency for the ‘neighbour’ Tn mutants compared with Tn_rotB. P-values for a 383 one-way ANOVA with Dunnetts multiple comparison test were 0.4197ns, 0.6682ns, 0.5882ns and 0.9999ns (left to 384 right). All values shown have been log transformed and show the means and standard deviation from the 3 385 replicate values.

386

387 Figure 8. Amplified specialised transduction fragments from precipitated phage DNA. A 0.6% agarose gel 388 showing the PCR product of a long-range amplification using Takara LA polymerase. Templates were as follows, 389 1. DNA extracted from precipitated phage capsids from an induced lysate of 8325-4 φ11 Tn_RotB, 2. DNA 390 extracted from precipitated phage capsids from an induced lysate of 8325-4 φ11 Tn_1866, 3. DNA extracted 391 from precipitated phage capsids from an induced lysate of 8325-4 φ11 Tn_1868, 4. Bacterial DNA from a 392 culture of 8325-4 φ11 Tn_rotB, 5. Bacterial DNA from a culture of 8325-4 φ11 Tn_1866, 6. Bacterial DNA from a 393 culture of 8325-4 φ11 Tn_1868, N. No template. The expected size for region φ11 to Tn_1866 was ~10.6 kb and 394 ~ 11.7 kb for region φ11 to Tn_1868. Three biological replicates were performed including induction, 395 precipitation, DNA extraction and PCR, and the above gel is representative.

20 bioRxiv preprint doi: https://doi.org/10.1101/2020.10.28.360172; this version posted October 29, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC 4.0 International license.

396 Regions containing the preferentially transduced transposon mutants are packaged as transductants at 397 higher rates than the rest of the genome

398 In addition to utilising our NTML transduction screen to identify transposon mutants that were preferentially 399 transduced, in a separate experiment we induced the ϕ11 in strain 8325-4 ϕ11 with mitomycin, and sequenced 400 the resultant, precipitated phage ϕ11 lysates to identify any bacterial DNA that had been packaged in phage 401 capsids. Fig. 9 shows the mapping of the reads to 8325-4, thus removing the majority of reads mapping to the 402 phage and allowing the lower frequency reads that mapped to the bacterial genome to be seen. The regions

403 up- and downstream of the ϕ11 attB (indicated by the red line) had higher numbers of mapped reads than the 404 other regions of the genome, supporting our screening data in which the identified mutants were clustered in 405 this same region. The pattern of the downstream bacterial DNA indicates that lateral transduction may be

406 responsible, with a gradual reduction of mapped reads from high numbers near to the attB, decreasing the 407 further away from the attachment site. Lateral transduction has been shown to taper off in this fashion, whilst 408 still facilitating transfer of ~ 200 kb of downstream bacterial DNA (14). Here we see slightly elevated read

409 mapping as far away as 798 kb from the attB site. The read alignments of the upstream bacterial DNA

410 encompass 117 kb of DNA upstream of the attB. These reads could represent the effect of specialised 411 transduction following the incorrect excision of the phage, albeit representing a larger region than one would 412 expect with a phage genome length of ~45 kb.

413

414

415 Figure 9. Mapping of lysate-derived DNA to the 8325-4 chromosome. The top graph shows read mapping to 416 part of the 8325-4 chromosome between 1.44 Mb and 2.05 Mb. The red line indicates the site of insertion of ϕ11 417 in the lysogen. Beneath this is the six frame translation (top three and bottom three lines) where the genes

21 bioRxiv preprint doi: https://doi.org/10.1101/2020.10.28.360172; this version posted October 29, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC 4.0 International license.

418 referred to in this study are represented by blue boxes, with the NTML gene numbers included. The central two 419 lines represent the forward and reverse DNA strands, with the concatenated assembly contigs identified with 420 alternating orange and brown boxes. The narrow peak of read mapping to the right of the insertion site is 421 caused by multiple mapping to a ribosomal RNA operon.

422

423 Discussion

424 This paper describes a high throughput method by which the mobile regions of bacterial genomes can be 425 identified. By screening pooled transposon mutants, we were able to identify different modes of phage-related 426 horizontal gene transfer, including phage transfer, lateral and specialised transduction. The method gave a 427 manageable number of transposon mutants to further investigate and could be applied to any strain for which 428 there is a transposon mutant library available. Prior to our screen we had anticipated obtaining two types of 429 mutants, namely those where the transposon was placed in the genome such that it could be transferred with 430 higher frequency, for example by lateral transduction, and mutants that due to inactivation of a bacterial gene 431 would lead to increased transduction efficacy. The results showed, however, that we primarily obtained 432 mutants belonging to the first category and, in addition, transposon elements that had inserted into mobile 433 genetic elements. Thus, for four of the resulting transposon mutants, the transposon element was inserted in 434 phages being endogenous to the JE2 strain, namely the USA300ϕSa2 and USA300ϕSa3, and these had been 435 transferred to the 8325-4 background. This phage transfer is not totally unexpected, since prophages can be 436 induced by an infecting phage (34) and the 8325-4 strain encodes attachment sites for these phages. However, 437 it is interesting to see both phages being successfully transferred and the resulting lysogens predominate their 438 pools. It suggests that the frequencies of prophage induction, during infection or by spontaneous release, 439 combined with lysogenization is greater than the transduction frequencies in this setup. One surprising transfer 440 event is that of the transposon inserted in gene SAUSA300_0810 from JE2 SaPI5, which is not present in the 441 8325-4 ϕ11 recipient, but the transposon was able to transfer to this strain anyway. On closer inspection, it 442 appears that this could be a case of generalised transduction, as there was no evidence of the SaPI5 integrase 443 that would indicate whole island transfer (data not shown) and NCBI Blast comparisons identified a region 444 close to the transposon insertion with homology to a chromosomal region in 8325-4. Further investigations 445 would have to be done to confirm exactly how and where this transduction took place, but it shows that 446 transfer can occur in a multitude of ways and that our methodology captures these.

22 bioRxiv preprint doi: https://doi.org/10.1101/2020.10.28.360172; this version posted October 29, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC 4.0 International license.

447 For two of the transposon mutants that were preferentially selected in our pooled screens, the transposon 448 elements were located downstream of the ϕ11 attachment site (SAUSA300_1849 and SAUSA300_1797). These 449 mutants displayed high transduction frequencies when transferred following induction of a ϕ11 lysogen, 450 compared to a control where the antibiotic resistance marker was located elsewhere in the bacterial genome, 451 and this effect was reduced when transfer was monitored following infection of the strains by ϕ11. These 452 results indicate that the increased transduction was dependent on induction of an integrated ϕ11, which is one 453 of the prerequisites for lateral transduction (15). Although both specialised and lateral transduction require 454 integration, specialised transduction is limited to regions immediately neighbouring the integrated prophage. 455 Therefore, though both specialised and lateral transduction could potentially transfer Tn_1849, which sits close 456 to the prophage (~8000 kb), Tn_1797 sits >39 kb downstream of the ϕ11 prophage, nearly a full phage genome 457 length away, so is unlikely to be transferred by specialised transduction. Furthermore, the original study 458 reporting lateral transduction found that a marker 5 kb downstream of the integrated phage was transferred 459 solely by lateral transduction and the researchers did not see any specialised particles containing this region, 460 suggesting all our downstream transfer can be attributed to lateral transduction (15, 17). This notion is 461 supported by the pattern of bacterial DNA sequences obtained from the sequencing of induced phage lysates 462 that indicates extensive packaging of bacterial DNA downstream of the integrated phage. Interestingly, this

463 data actually implicates lateral transduction at an even greater distance downstream of the attB site then has 464 previously been suggested (14), with slightly elevated reads mapping up to 798 kb downstream.

465 It has been shown that temperate phages are capable of both generalised and specialised transduction (14, 17, 466 20) but until now, specialised transduction had not been reported in S. aureus. However, for the transposon 467 mutants Tn_1866 and Tn_1868 identified in our screen, the transposon element was located immediately 468 upstream of the ϕ11 integration site at a distance of 7.6 and 8.7 kb, respectively, thus placing it comfortably 469 within a region where specialised transduction could be expected to occur. Further, the increased transduction 470 frequencies of Tn_1866 and Tn_1868 were eliminated when lysates were created by infection rather than 471 induction. Equally, they should not be transferred by lateral transduction, which has been shown to only occur 472 downstream of the phage pac site, in a unidirectional manner defined by the phage packaging machinery (14). 473 This left specialised transduction as the remaining, feasible explanation. Indeed, we were able to identify PCR 474 product containing both ϕ11 and the transposon using precipitated phage lysate from the strains with Tn_1866

475 and Tn_1868. This indicated that some phage capsids contained both the attL region of the phage together with 476 the upstream flanking region up to 8.7 kb away. Taken together with the reduced transduction frequency

23 bioRxiv preprint doi: https://doi.org/10.1101/2020.10.28.360172; this version posted October 29, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC 4.0 International license.

477 following infection rather than induction, this supports that the upstream genes are transferring through 478 specialised transduction.

479 In the transfer of the transposon elements located upstream of the ϕ11 attachment site, we observed ~22-fold 480 higher transduction frequencies compared to our control, which would suggest that, although aberrant 481 excision is generally accepted as a rare event, such particles have a high rate of successful transfer. This could 482 be due to their exploitation of the phage integrase for recombination into the recipient host DNA. On the other 483 hand, it could also be that aberrant excision happens at a higher frequency than expected for ϕ11. Potentially, 484 the in situ replication that has been shown to occur both upstream and downstream of the prophage when 485 ϕ11 has been induced, but has not yet excised, could be contributing to the rate of specialised 486 transduction(17). However, further work is needed to fully establish the specialised packaging mechanism. In 487 either case, the occurrence of specialised transduction is supported by DNA sequencing of the phage particles, 488 which shows elevated levels of this bacterial upstream region up to 117 kb away from the phage in the 489 precipitated phage lysate. Given that specialised transduction is thought to rely at least in part on the phage 490 integrase for recombination, this could contribute to why we only see upstream genes transferred in this way 491 (17). If prophage ϕ11 went through an aberrant excision event involving the downstream bacterial flank near

492 the attR, it would likely lose the integrase in the process, which could make successful recombination into the 493 recipient less likely. However, given that in our set up the wild-type prophage is present in the recipient, there 494 is likely additional factors contributing to the lack of downstream specialised transduction. Kwoh and Kemper 495 discussed the possible variations of specialised transduction in Salmonella typhimurium at length, but there 496 could be differences in specialised transduction between this organism and S. aureus (17).

497 Another interesting aspect of our data is the slightly higher transduction frequencies observed upon infection 498 seen for the Tn mutations located downstream of the ϕ11, which we predict are transferred by lateral 499 transduction. Such an effect could suggest the integration of a small number of phages during the infection 500 cycle, as proposed in Salmonella phage P22 (35). A ‘lyso-lysis’ effect has been shown in E. coli where cells 501 infected by multiple phages at once displayed simultaneous phage lysogenic integration into the genome and 502 entry into the lytic cycle (15). Potentially this is occurring during the ϕ11 infection cycle and leading to low 503 levels of specialised and lateral transduction being possible. Current investigations of the effect of integration 504 during the lytic cycle on transduction are ongoing.

505 A limitation of the screening set up described here is that we are analysing the mobility of genes related both 506 to phage infection and prophage induction. Future screens could be optimized to specifically address individual

24 bioRxiv preprint doi: https://doi.org/10.1101/2020.10.28.360172; this version posted October 29, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC 4.0 International license.

507 aspects of DNA transfer. Another consideration is the co-selection for transposon mutants with greater fitness, 508 that arise because of the necessary incubation periods during the screening. This may explain why within our 509 pooled mutants we identified a limited number of transposon mutants from the highly transduced regions, 510 even though others were present in the pool. The transposon mutants identified in the screen may well 511 represent those of higher fitness relative to the other highly transduced transposon mutants from that pool.

512 In summary, this study acts as a proof of concept for using transposon mutant libraries to investigate which 513 regions of the bacterial genome are highly mobile. We show phage transfer, generalised, specialised, and 514 lateral transduction between staphylococcal strains. While the set up employed here was used to identify 515 highly transduced genes and chromosomal locations, it may in fact also reflect what happens over time in the 516 real world. Release of ϕ11 from lysogens promotes acquisition of bacterial DNA by auto-transduction, and 517 resulting transductants will as lysogens be able to repeat the process (27). Horizontal gene transfer underpins 518 the evolution and adaptability of S. aureus and approaches such as the one adopted here will help us increase 519 our understanding of which regions are transferred and how.

520

521 References

522 1. Appelbaum PC. 2006. The emergence of vancomycin-intermediate and vancomycin-resistant 523 Staphylococcus aureus. Clinical Microbiology and Infection 12:16–23.

524 2. DeLeo FR, Otto M, Kreiswirth BN, Chambers HF. 2010. Community-associated meticillin-resistant 525 Staphylococcus aureus. The Lancet. Elsevier.

526 3. Goerke C, Wolz C. 2004. Regulatory and genomic plasticity of Staphylococcus aureus during persistent 527 colonization and infection. International Journal of Medical Microbiology. Elsevier GmbH.

528 4. Lindsay JA. 2014. Staphylococcus aureus genomics and the impact of horizontal gene transfer. 529 International Journal of Medical Microbiology. Urban & Fischer.

530 5. Xia G, Wolz C. 2014. Phages of Staphylococcus aureus and their impact on host evolution. Infection, 531 Genetics and Evolution 21:593–601.

532 6. Deghorain M, van Melderen L. 2012. The Staphylococci Phages Family: An Overview. Viruses 4:3316– 533 3335.

534 7. Iandolo JJ, Worrell V, Groicher KH, Qian Y, Tian R, Kenton S, Dorman A, Ji H, Lin S, Loh P, Qi S, Zhu H, Roe 535 BA. 2002. Comparative analysis of the genomes of the temperate bacteriophages φ11, φ12 and φ13 of 536 Staphylococcus aureus 8325. Gene 289:109–118.

25 bioRxiv preprint doi: https://doi.org/10.1101/2020.10.28.360172; this version posted October 29, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC 4.0 International license.

537 8. Gillaspy AF, Worrell V, Orvis J, Roe BA, Dyer DW, Iandolo JJ. 2014. The Staphylococcus aureus NCTC 8325 538 Genome, p. 381–412. In Gram-Positive Pathogens. ASM Press, Washington, DC, USA.

539 9. Lee CY, Iandolo JJ. 1988. Structural analysis of staphylococcal bacteriophage phi 11 attachment sites. 540 Journal of bacteriology.

541 10. Sternberg N, Coulby J. 1990. Cleavage of the bacteriophage P1 packaging site (pac) is regulated by 542 adenine methylation (DNA adenine methyltransferase/methylation/phage DNA packaging) Proc. Natl. 543 Acad. Sci. USA.

544 11. Jackson EN, Jackson DA, Deans RJ. 1978. EcoRI analysis of bacteriophage P22 DNA packaging. Journal of 545 Molecular Biology 118:365–388.

546 12. Lennox ES. 1955. Transduction of linked genetic characters of the host by bacteriophage P1. Virology 547 1:190–206.

548 13. ZINDER ND, LEDERBERG J. 1952. Genetic exchange in Salmonella. Journal of bacteriology 64:679–699.

549 14. Chen J, Quiles-Puchalt N, Chiang YN, Bacigalupe R, Fillol-Salom A, Chee MSJ, Fitzgerald JR, Penadés JR. 550 2018. Genome hypermobility by lateral transduction. Science (New York, NY) 362:207–212.

551 15. Kwoh DY, Kemper J. 1978. Bacteriophage P22-mediated specialized transduction in Salmonella 552 typhimurium: high frequency of aberrant prophage excision. Journal of Virology 27:519–534.

553 16. Zahler SA, Korman RZ, Rosenthal R, Hemphill H. 1977. Bacillus subtilis bacteriophage SPβ: localization of 554 the prophage attachment site, and specialized transduction. Journal of Bacteriology 129:556–558.

555 17. Kwoh DY, Kemper J. 1978. Bacteriophage P22-mediated specialized transduction in Salmonella 556 typhimurium: identification of different types of specialized transducing particles. Journal of Virology 557 27:535–550.

558 18. Mackey CJ, Zahler SA. 1982. Insertion of bacteriophage SP beta into the citF gene of Bacillus subtilis and 559 specialized transduction of the ilvBC-leu genes. Journal of bacteriology 151:1222–9.

560 19. Cavenagh MM, Miller R v. 1986. Specialized transduction of Pseudomonas aeruginosa PAO by 561 bacteriophage D3. Journal of Bacteriology 165:448–452.

562 20. Hoppe I, Roth J. SPECIALIZED TRANSDUCING PHAGES DERIVED FROM SALMONELLA PHAGE P22.

563 21. Porter RD, Lark MW, Low KB. 1981. Specialized transduction with lambda plac5: dependence on recA 564 and on configuration of lac and att lambda. Journal of virology 38:497–503.

565 22. Campos J, Martínez E, Marrero K, Silva Y, Rodríguez BL, Suzarte E, Ledón T, Fando R. 2003. Novel Type 566 of Specialized Transduction for CTX or Its Satellite Phage RS1 Mediated by Filamentous Phage VGJ in 567 Vibrio cholerae. JOURNAL OF BACTERIOLOGY 185:7231–7240.

568 23. Fink PS, Zahler SA. 1982. Specialized transduction of the ilvD-thyB-ilvA region mediated by Bacillus 569 subtilis bacteriophage SP beta. Journal of bacteriology 150:1274–9.

26 bioRxiv preprint doi: https://doi.org/10.1101/2020.10.28.360172; this version posted October 29, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC 4.0 International license.

570 24. Porter RD, Welliver RA, Witkowski TA. 1982. Specialized transduction with lambda plac5: dependence 571 on recB. Journal of Bacteriology 150:1485.

572 25. Kleiner M, Bushnell B, Sanderson KE, Hooper L v., Duerkop BA. 2020. Microbial DNA on the move: 573 sequencing based detection and analysis of transduced DNA in pure cultures and microbial 574 communities. BioRxiv 2020.01.15.908442.

575 26. Fey PD, Endres JL, Yajjala VK, Widhelm TJ, Boissy RJ, Bose JL, Bayles KW. 2013. A genetic resource for 576 rapid and comprehensive phenotype screening of nonessential Staphylococcus aureus genes. mBio 4.

577 27. Haaber J, Leisner JJ, Cohn MT, Catalan-Moreno A, Nielsen JB, Westh H, Penadés JR, Ingmer H. 2016. 578 Bacterial viruses enable their host to acquire antibiotic resistance genes from neighbouring cells. Nature 579 Communications 7:1–8.

580 28. Bae T, Class EM, Schneewind O, Missiakas D. 2007. Generating a collection of insertion mutations in the 581 Staphylococcus aureus genome using bursa aurealis. Methods in Molecular Biology 416:103–116.

582 29. Page AJ, de Silva N, Hunt M, Quail MA, Parkhill J, Harris SR, Otto TD, Keane JA. 2016. Robust high- 583 throughput prokaryote de novo assembly and improvement pipeline for Illumina data. Microbial 584 genomics 2:e000083.

585 30. SMALT – Wellcome Sanger Institute.

586 31. Carver T, Harris SR, Berriman M, Parkhill J, Mcquillan JA. 2012. Artemis: an integrated platform for 587 visualization and analysis of high-throughput sequence-based experimental data 28:464–469.

588 32. Goerke C, Pantucek R, Holtfreter S, Schulte B, Zink M, Grumann D, Bröker BM, Doskar J, Wolz C. 2009. 589 Diversity of prophages in dominant Staphylococcus aureus clonal lineages. Journal of Bacteriology 590 191:3462–3468.

591 33. Chiang YN, Penadés JR, Chen J. 2019. Genetic transduction by phages and chromosomal islands: The 592 new and noncanonical. PLOS Pathogens 15:e1007878.

593 34. Campoy S, Hervàs A, Busquets N, Erill I, Teixidó L, Barbé J. 2006. Induction of the SOS response by 594 bacteriophage lytic development in Salmonella enterica. Virology 351:360–367.

595 35. Shao Q, Trinh JT, McIntosh CS, Christenson B, Balázsi G, Zeng L. 2017. Lysis-lysogeny coexistence: 596 prophage integration during lytic development. MicrobiologyOpen 6.

597

598